Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
42.1F1 Score
7
Apr 21, 2026
32.7EM (Exact Match)
7
Apr 20, 2026
7.8Exact Match (EM)
7
Apr 20, 2026
59.3EM
7
Apr 20, 2026
34.3EM
7
Apr 20, 2026
35.6EM
7
Apr 20, 2026
49.9Time F1
7
Apr 16, 2026
26.8Few-Shot Accuracy
7
Apr 14, 2026
26.62F1 Score
7
Mar 23, 2026
77.5Business Ethics Accuracy
7
Mar 20, 2026
95.6MCQ Accuracy
7
Mar 18, 2026
96.8MCQ Accuracy
7
Mar 18, 2026
98.4MCQ Accuracy
7
Mar 18, 2026
81.8Normalized Log Accuracy (MMLU)
7
Mar 20, 2026
64.2Exact Match (EM)
7
Mar 17, 2026
93.12ScienceQA Accuracy
7
Mar 13, 2026
0.88Accuracy
7
Mar 11, 2026
65.31Precision
7
Mar 5, 2026
71.7Precision
7
Mar 5, 2026
0.071ECE
7
Mar 4, 2026
26.99F1 Score
7
Feb 27, 2026
24.17F1 Score
7
Feb 27, 2026
23.59F1 Score
7
Feb 27, 2026
28.76F1 Score
7
Feb 27, 2026
24.96F1 Score
7
Feb 27, 2026